Endüstriyel Sistemlerde Arkaplan Çıkarımı Tabanlı Hareketli Nesne Tespiti ve Sayılması için Yeni Bir Yaklaşım

Abstract

Bilgisayarlı görme ve görüntü işleme yaklaşımları günümüzde endüstriyel kontrol sistemlerinde oldukça önemli bir yer tutmaktadır. Özellikle kalite kontrol sistemlerinde kullanılan bilgisayarlı görme teknolojisi bir üretim hattında üretilen ürünlerin hızlı, sorunsuz ve doğru bir şekilde kontrol edilmesi açısından oldukça büyük öneme sahiptir. Klasik metotlarla yapılan kontrollerin getireceği problemler göz önüne alındığında bilgisayarlı görme kavramının ehemmiyeti daha net bir şekilde anlaşılacaktır. Bu çalışmada bilgisayarlı görme tabanlı kalite kontrolü için yeni bir metot önerilmiştir. Önerilen metot bir üretim hattından geçen ürünlerin görüntülerini kaydeder. Üretim bandından geçen bu ürünler saniyede 60 görüntü karesi (fps) hızına sahip bir kamera aracılığıyla kaydedilmiştir. Önerilen yaklaşımda alınan görüntüler öncelikle bazı morfolojik işlemlerden geçirilerek ürünlerin detaylarının net bir şekilde ortaya çıkarılması sağlanmıştır. Daha sonra Gaussian karışım modeli ile arkaplanı çıkarılarak hareket eden nesneler tespit edilmiştir. Daha sonra belirlenen bölgeden nesnelerin geçip geçmediği kontrol edilerek saydırma işlemi yapılmıştır. Birbirine yakın veya üst üste binmiş olan nesnelerin doğru şekilde saydırılması için Convex hull yöntemi ile nesnelerin kontur’ü çizdirilmiş ve nesnelerin alanına göre alandan çoklu nesne geçip geçmediği tespit edilmiştir. Önerilen bu yaklaşım ile yüksek hıza sahip üretim bantlarında geçen ürünlerin eksiksiz ve sorunsuz bir şekilde sayımı gerçekleştirilmiş olup, elde edilen deneysel sonuçlar ile algoritmanın etkili ve doğru sonuçlar verdiği gözlemlenmiştir.

Keywords

• Bilgisayarlı Görme,Nesne Sayma,Yüksek Hızlı Görüntü İşleme,Endüstriyel Uygulamala

A New Approach for Background Subraction Based Moving Object Detection and Counting in Industrial System

Abstract

In recent years, computer vision and image processing approaches hold a very important place in industrial control systems. Especially, the computer vision technique used in the quality control of products in a production line has a great importance in terms of controlling of products fast, smoothly and correctly. When it is considered problems brought up by the inspections carried out by conventional methods, the importance of computer vision concept will be understood more clearly. In this study, a new method was proposed for computer vision based quality control. The proposed method records images of products passed on a production line as a video. These products were counted by using high speed image processing techniques. Products passed on a production line were recorded by using a camera with 60 frame per second (fps). In the proposed method, the images taken in the proposed approach are firstly subjected to some morphological operations to reveal the details of the products clearly. Then, moving objects were detected by removing the background with Gaussian mixture model. Then, it is checked whether or not the objects in the determined area have passed, and the process of counting is performed. The contours of the objects are drawn with Convex hull method and it is determined whether multiple objects pass over the area according to the area of the objects. The counting of the products in high speed production line was performed successfully and completely and the experimental results show that the algorithm is effective and accurate results. 

Keywords

• Computer Vision,Object Counting,High Speed Image Processing,Industrial Applications.

References

1. [1] Lee K. H., Park T. H. Image segmentation of UV pattern for automatic paper-money inspection, 11th International Conference on Control Automation Robotics and Vision (ICARCV), 1175-1180, 2010.
2. [2] Lee J. H., Lee J. M., Kim H. J., Moon Y. S. Machine vision system for automatic inspection of bridges, Congress on Image and Signal Processing (CISP), 3, 363366, 2008.
3. [3] Chen Y. R., Chao K., Kim M. S. Machine vision technology for agricultural applications, Computers and Electronics in Agriculture, 36, 173-191, 2002.
4. [4] Yoshino K., Miwa M., Kanamaru A., Kanai N. An automatic through-hole inspection system by analyzing laser diffraction pattern, Proceedings of Annual Conference (SICE), 2155-2160, 2010.
5. [5] Duan G., Chen Y. W., Sakekawa T. Automatic optical inspection of micro drill bit in printed circuit board manufacturing based on pattern classification, Instrumentation and Measurement Technology Conference Proceedings (IMTC), 279-283, 2008.
6. [6] Khan U. S., Iqbal J., Khan M. A. Automatic inspection system using machine vision, 34th Applied Imagery and Pattern Recognition Workshop (AIPR), 212-217, 2005.
7. [7] Zhou W., Fei M., Zhou H., Li K. A sparse representation based fast detection method for surface defect detection of bottle caps, Neurocomputing, 123, 406-414, 2014.
8. [8] Yang S. W., Lin C. S., Lin S. K., Tseng Y. C. Automatic inspection system for defects of printed art tile based on texture feature analysis, Instrumentation Science and Technology, 42, 59-71, 2013.

9. [9] Yazdi L., Prabuwono A. S., Golkar E. Feature extraction algorithm for fill level and cap inspection in bottling machine, International Conference on Pattern Analysis and Intelligent Robotics (ICPAIR), 1, 47-52, 2011.
10. [10] Brosnan T., Sun D. W. Inspection and grading of agricultural and food products by computer vision systems-a review, Computers and Electronics in Agriculture, 36, 193-213, 2002.
11. [11] Torregrosa A., Albert F., Aleixos N., Ortiz C., Blasco J. Analysis of the detachment of citrus fruits by vibration using artificial vision, Biosystems Engineering, 119, 112, 2014.
12. [12] Halfawy M. R., Hengmeechai J. Optical flow techniques for estimation of camera motion parameters in sewer closed circuit television inspection videos, Automation in Construction, 38, 39-45, 2014.
13. [13] Stojanovic R., Mitropulos P., Koulamas C., Karayiannis Y., Koubias S., Papadopoulos G. Real-time vision based system for textile fabric inspection, Real-Time Imaging, 7, 507-518, 2001.
14. [14] Cho C. S., Chung B. M., Park M. J. Development of realtime vision-based fabric inspection system, IEEE Transactions on Industrial Electronics, 52, 1073-1079, 2005.
15. [15] Kumar A. Computer vision based fabric defect detection: a survey, IEEE Transactions on Industrial Electronics, 55, 348-363, 2008.
16. [16] Jia H., Murphey Y. L., Shi J., Chang T. S. An intelligent real-time vision system for surface defect detection, Proceedings of the 17th International Conference on Pattern Recognition (ICPR), 3, 239-242, 2004.
17. [17] Ge X. The design of a global shutter CMOS image sensor in 110 nm technology, Master of Science Thesis, Delft University of Technology, 2012.
18. [18] Lim S. H. Video-processing applications of high speed cmos image sensors, The Degree of Doctor of Philosophy, Stanford University, 2003.
19. [19] Palakodety A. CMOS active pixel sensors for digital cameras: current state of the art, The Degree of Master of Science, University of North Texas, 2007.
20. [20] Santur, Y., Karaköse, M., Akın, E. Learning Based Experimental Approach for Condition Monitoring Using Laser Cameras in Railway Tracks, International Journal of Applied Mathematics, Electronics and Computers (IJAMEC), 4, 1-5, 2016.
21. [21] Yetis H., Baygin M., Karaköse M. A New Micro Genetic Algorithm Based Image Stitching Approach for Camera Arrays at Production Lines, The 5th International Conference on Manufacturing Engineering and Process (ICMEP 2016), 25-27 May, 2016.
22. [22] Karaköse M., Yaman O., Aydin I., Karakose E., RealTime Condition Monitoring Approach of PantographCatenary System Using FPGA, 14th IEEE International Conference on Industrial Informatics (IEEE INDIN 2016), Futuroscope-Poitiers, France, 18-21 July 2016.
23. [23] Aydin I., Karakose E., Karaköse M., Gençoğlu M.T., Akın E., A New Computer Vision Approach for Active Pantograph Control, IEEE International Symposium on Innovations in Intelligent Systems and Applications (IEEE INISTA 2013), Albena, Bulgaria, 2013.